INFT216代写、代做BondTelco留学生、代写Java/c++编程语言、代做Python

日期：2019-04-06 09:05

INFT216 - Data Science 191

Assignment 4: Finding customer clusters at BondTelco

Assessment Value: 10%

Due Date: End of Week 12, Sunday 7th April 5pm (submit via iLearn)

Previously, management have been quite directed about what they were looking for and what they

wanted to predict. Now, they are asking you what kind of things can be ‘discovered’ from the data.

In particular they are interested in whether there are any kind of natural groupings that exist within

their customer base.

You have access to the same data as Assignment 1.

COLLEGE : Is the customer college educated?

INCOME : Annual income

OVERAGE : Average overcharges per month

LEFTOVER : Average % leftover minutes per month

HOUSE : Value of dwelling (from census tract)

HANDSET_PRICE : Cost of phone

OVER_15MINS_CALLS_PER_MONTH : Average number of long (>15 mins) calls

per month

AVERAGE_CALL_DURATION : Average call duration

REPORTED_SATISFACTION : Reported level of satisfaction

REPORTED_USAGE_LEVEL : Self-reported usage level

CONSIDERING_CHANGE_OF_PLAN : Was customer considering changing

his/her plan?

LEAVE : Whether customer left or stayed

Your goal is to find and explain any one natural grouping you find within the data. You only need to

concentrate on finding one way to group customers, and then explain that grouping in business

terms.

Deliverables:

Your final deliverables will be 2 PDF files, both produced by the same .Rmd script (with different

code chunk options). You must submit:

PDF1 - all code and results shown (like you would share with a colleague on the Data Science team)

PDF2 - only show those things necessary to help support management decision making (this is the

one you send to management!)

These will both be submitted online through iLearn.

Guidance:

Document

o Focus on a good document structure and layout (revisit week 2 on repeatable

research)

o Hint: Think about the headings in the document you produce

Focus on letting the visualizations do the talking. Only include explanatory text where it is

really necessary… although you should remember that management do not really

understand data science, so you will need to find a tradeoff between understandability and

verbosity. Verbose assignments will be penalized.

o You will need to use visualizations

o You will need to explain at least 1 of the clusters - and it must be useful explained

from a business perspective

Note:

As is the case with all assignments I set, if you do the minimum (correctly), then you will receive half

marks. Additional marks are awarded for those assignments where you have clearly put in

additional thought, whether it be in visualization, modelling, succinctness, or coding elegance.